Summary of study 1: Clinical features of chronic pain with neuropathic characteristics: a

This study aimed to assess the severity of self-reported neuropathic symptoms in different syndromes of chronic pain and to group the syndromes according to the severity of their

neuropathic symptoms. These groups were compared regarding pain variables like intensity and chronicity, as well as pain-related disability, length of hospital stay, pain history and depressive symptoms. The predictive strength of the above mentioned variables regarding the severity of neuropathic symptoms was to be analysed.

Four hundred patients enrolled for multidisciplinary pain treatment were considered for inclusion in the study. The criteria for inclusion were: an age over 18 years and a diagnosis of chronic pain according to ICD-10 criteria (F45.41 or R52.1-2, International Statistical Classification of Diseases and Related Health Problems, 2012). The classification of pain syndromes was based on IASP Taxonomy (Turk & Rudy, 1987), using the well validated

“Multiaxial Pain Classification System- Somatic Dimension” (MASK-S, Hildebrandt, Pfingsten, Maier, & Klinger, 1992; Klinger, Hasenbring, Pfingsten, Hürter, Maier, & Hildebrandt, 2000). A total of five primary diagnoses (i.e., headache, spinal column pain (mainly including back pain), musculoskeletal pain, typical neuropathic pain, postsurgical pain) were assigned. Because of the low numbers of patients with facial pain, ischemic pain, visceral pain and somatic unclassifiable pain, these diagnostic groups were collapsed into a single category, i.e., other pain.

The severity of neuropathic symptoms in each diagnostic group was assessed by the pain DETECT questionnaire (PDQ, Freynhagen et al., 2006a). One-way analyses of variance (ANOVAs) and post-hoc Tukey's tests were conducted to explore differences between the diagnostic groups regarding the severity of neuropathic symptoms (PDQ score).

According to Freynhagen et al. (2006a), the above mentioned diagnostic groups were recategorised on the basis of the severity of their neuropathic symptoms: Neuropathic (NP), non-Neuropathic (no-NP) and Unclear (UC) groups. ANOVAs and post-hoc Turkey's tests were performed in order to test the differences between the three groups (NP, no-NP, UC) regarding pain intensity, pain chronicity, disability, length of hospital stay, pain history and depressive symptoms. Univariate logistic regressions were applied to determine the association of these variables with the severity of neuropathic symptoms. The variables with a significant association to the severity of neuropathic symptoms were fed into a hierarchical multiple regression analysis.

Our findings demonstrated the presence of distinct self-reported neuropathic symptoms in 37% of all patients included. The validity of the PDQ score was supported by the fact that patients who had been clinically diagnosed with “typical neuropathic pain” scored highest on the PDQ (M=17.79, SD=6.38; F (5, 394)=2.26; p=0.04), being significantly higher than in three of the other groups, i.e., the “spinal column pain”, “headache” and “other pain” groups, with the exception of “musculoskeletal pain” and “postsurgical pain”.

A high PDQ score in the musculoskeletal pain category in the present study may result from a high percentage of fibromyalgia patients (60%, PDQ scorefibromyalgia = 17.70) in this group in our sample. As mentioned earlier, similarities in neuropathic symptoms between fibromyalgia and “typical neuropathic pain” may be explained by pathogenic mechanisms such as impaired small fibre function and a dysfunction of the endogenous systems modulating afferent activity in fibromyalgia (Bradley et al., 2002; Kosek et al., 1996; Uceyler et al., 2013). Although there is no

evidence of nerve injury in the majority of fibromyalgia patients and thus not fulfilling the criteria for typical neuropathic pain, the PDQ cannot distinguish fibromyalgia from neuropathic pain disorders on the basis of the neuropathic symptom score. This failing differentiation is a clear disadvantage for the choice of adequate treatment strategies. Furthermore, since some previous studies (e.g., Gormsen, Rosenberg, Bach, & Jensen, 2010; Koroschetz et al., 2011) have documented higher levels of pain intensity, depression and anxiety in fibromyalgia patients compared to neuropathic pain patients, this finding should be considered when interpreting the association of the severity of neuropathic symptoms with the intensity of pain and psychological distress in samples of chronic pain patients while not considering the proportion of fibromyalgia patients in the sample.

We also found a high severity of neuropathic symptoms in “postsurgical pain” not significantly differing from “typical neuropathic pain”. This finding provides some support for the assertion that a large component of persistent pain after surgery is defined by somatosensory symptoms defining as neuropathic pain (e.g., Kehlet, Jensen, & Woolf, 2006; Shaladi et al., 2009; Shipton, 2008).

Contrasting this result, a significantly lower severity of neuropathic symptoms was found in

“spinal column pain” compared to “typical neuropathic pain”. According to Freynhagen et al.

(2006b), only one-third of chronic back pain patients reported three or more neuropathic pain symptoms. This indicates that, although in a minor proportion of patients with back pain (e.g.

radiculopathy) neuropathic mechanisms play a distinct role regarding their pain (Mahn et al., 2011), this is not the case in the majority of these patients and also clearly less neuropathic symptoms are reported.

Altogether, these results obviously demonstrate a spectrum of expression of neuropathic symptoms in different syndromes of chronic pain that challenges the dichotomous classification of chronic pain as nociceptive or neuropathic pain. Moreover, our findings lend support to the notion that, although diverse types of pain are distinct in their aetiology, they share some

similarities in the underlying pathophysiological mechanisms of pain generation (Costigan, Scholz, & Woolf, 2009).

Our findings, based on a clinical sample of patients with different types of chronic pain, demonstrated a higher level of pain intensity and depressive symptoms in the NP and UC groups compared with the no-NP group. ANOVAs and post-hoc tests demonstrated also significant differences between each of the analysed groups (NP, UC, no-NP) regarding pain chronicity and disability. The length of hospital stay in the NP group was significantly higher than in the no-NP group (see original article, Table 2). No group differences were found regarding pain history.

The results are consistent with findings in population-based studies (e.g., Attal et al., 2011;

Freynhagen et al., 2006) but in contrast to Daniel et al. (2008), who did not find any differences regarding pain intensity, depressive/anxiety symptoms and physical function between patients with postherpetic neuropathic pain and nociceptive back pain. The following possible reasons for this discrepancy should be taken into account. First, the clinical characteristics of diverse types of chronic pain might influence the observed association. While in our study a high percentage of fibromyalgia patients (61%) was in the NP group, this was not the case in described clinical studies comparing typical neuropathic pain and nociceptive pain syndromes. Past research has documented the highest level of pain intensity, disability, depression and anxiety among fibromyalgia patients compared to other chronic pain patients, even those with neuropathic pain (e.g., Gormsen et al., 2010; Koroschetz et al., 2011), whatever the cause for this phenomenon might be. Thus, it is necessary to examine the association of neuropathic symptoms with the pain and psychological factors in specific diagnostic groups like fibromyalgia, separately.

Second, the possibility of response bias should be considered. As noted before, not only does the character of experienced pain associated with biological mechanisms determine the symptom report, but also cognitive, emotional and behavioural factors. The contribution of the

psychological factors to the symptom report is more likely when patients do not have any identified pathology (Deary et al., 2007) or previous experience or knowledge about the

questioned symptoms. Based on clinical experience, the questioned symptoms in Pain DETECT Questionnaire like “is cold or heat (bath water) in this area occasionally painful?” or “do you have sudden pain attacks in the area of your pain, like electric shocks” were unfamiliar for many patients who had no previous experience about them. It is possible that some of these patients having a high level of negative affectivity (e.g. depression) and negative cognitive appraisals (e.g. pain catastrophizing) (mis)interpret and (mis)attribute rather inconspicuous sensations in the manner of the questioned neuropathic symptoms, inflating correlations among neuropathic symptoms and cognitive-emotional factors (e.g. depression). To gain a more precise insight regarding the association of neuropathic symptoms with pain and psychological factors, it should be analysed in a sample of patients who have been medically diagnosed with neuropathic

components of pain (e.g. typical neuropathic pain syndromes).

In summary, the results of study 1 support the questioning of the dichotomous classification of chronic pain since a high severity of neuropathic symptoms was found in other diagnoses, particularly, in fibromyalgia (as a dysfunctional pain condition) and postsurgical pain (as a mixed pain syndrome). Chronic pain patients who scored high on self-reported neuropathic symptoms also reported high levels of pain intensity, pain chronicity, depression and functional disability. These findings provide some support for the common assumption that the neuropathic quality of pain is experienced and reported to be more intense and distressing than the pain without this specific quality. Whether this phenomenon is mainly based on the biological mechanisms of pain (neuropathic/nociceptive) or on other factors of pain processing so far remain obscure.

3.2 Summary of study 2: Neuropathic sensory symptoms: Association with pain and